8results about How to "Efficient and accurate judgment" patented technology

The invention discloses a recommendation optimization method based on multi-modal feature clustering, and aims to improve diversity and user experience of a recommendation system. According to the method, firstly, features are extracted from a commodity title and a cover picture through a text and visual feature extraction model, and text and visual feature representation is obtained; then, the features are subjected to feature alignment and standardization processing to ensure the consistency of different modal features, and then the features are fused into a multi-modal embedded vector to serve as unified feature representation of commodities. And then, clustering is carried out on the fused multi-modal embedded vectors by using a K-Means clustering algorithm, so that clustering identifiers of the commodities can be formed. In the recommendation process, the system monitors the clustering identifiers of the displayed commodities, adjusts the recommendation strategy according to the identifiers, and preferentially recommends the commodities of different clusters from the displayed commodities, so that excessive distribution of similar contents is avoided, and the diversity of the recommended contents is ensured. According to the method, by integrating multi-modal information, the performance of a recommendation system is optimized, and the problem of excessive distribution is effectively solved.

The invention discloses a method for controlling a flash memory module, a related flash memory controller and a memory device. In the method, the flash memory module comprises a plurality of crystal grains, and the method comprises the following steps: if abnormal power failure occurs before the flash memory module is powered on, determining a super block which is written into the flash memory module at the last time before the flash memory module is powered on, and the super block comprises a plurality of blocks which are respectively positioned in the plurality of crystal grains; determining the last written data page of the super block; determining a check range of the super block according to the last written data page of the super block; and reading at least one part of data pages in the checking range to judge that the last data page of each block in the super block can be successfully read.

The invention discloses a method and a system for supervising internet data processing, and relates to the technical field of internet supervising, the method comprises the steps of flow data capture, data doubt judgment, problem data coping and early warning prompt, real-time supervision is executed on each internet transmission data packet, each doubt data packet is captured, and each doubt data packet is sent to the internet supervising module; the method comprises the following steps: receiving doubt data packets, executing corresponding analysis and reduction on the doubt data packets, judging whether the doubt data packets are malicious data packets, extracting features corresponding to the malicious data packets, judging to obtain malicious types corresponding to the malicious data packets, and analyzing to obtain countermeasures of the malicious types corresponding to the malicious data packets. According to the method, the security of the Internet data in acquisition and transmission is guaranteed, the application security corresponding to the Internet data is reduced, and meanwhile, the data risk is quickly and efficiently found based on judgment on the malicious data packet corresponding to the Internet data, so that timely data security risk measure prevention is realized.

The invention belongs to the technical field of data detection and identification, and discloses a data forgery detection method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining to-be-detected data; extracting data features of the to-be-detected data; comparing the time sequence feature with a time sequence of to-be-detected data in the time sequences of the multiple pieces of data to obtain a first comparison result; comparing the content characteristics with index contents of the to-be-detected data and / or session contents in a session file corresponding to the to-be-detected data to obtain a second comparison result; inputting the grammatical features into a pre-trained grammatical feature classification model to obtain feature types to which the grammatical features belong; and determining the to-be-detected data as forged data under the condition that the forged data condition is met. Therefore, by combining the extracted multi-dimensional data features with the grammatical feature classification model, the data can be efficiently and accurately subjected to forgery detection.

The application relates to a user natural gas safety detection system, a method and a storage medium, and the method comprises the following steps: reading user information of a user natural gas meter, wherein the user information comprises a user ID and a gas record; dividing the user type of the user ID based on the gas record; generating a trigger signal with different detection frequencies based on the user type; obtaining gas information corresponding to the user ID based on the trigger signal to update the gas record; judging whether a natural gas pipeline exists leakage based on the gas information, and issuing an alarm information when the leakage judgment result is that leakage exists, wherein the alarm information comprises the user ID; issuing a valve closing action instruction to the natural gas meter corresponding to the user ID based on the alarm information, and recording the alarm information in the gas record. The application has the effects of improving the safety of a household natural gas pipeline and realizing global natural gas safety detection and protection for community users.

The invention discloses a service request scene judgment method and device, computer equipment, a medium and a product. The method comprises the steps of obtaining to-be-processed request data input by a user; based on a preset white rule base, judging whether a current request is a normal service request or not according to static content characteristics in the to-be-processed request data; and if not, judging whether the current request is a normal service request or not according to the dynamic behavior characteristics in the to-be-processed request data based on a preset behavior rule base. According to the technical scheme, the normal service requests are quickly filtered out through the preset white rule base, so that the overall judgment efficiency is improved, then the complex requests which are not filtered out through the preset white rule base are further judged and processed from the dynamic behavior characteristics through the preset behavior rule base, and thus the overall judgment accuracy is improved. The two rule libraries supplement each other, the judgment efficiency and accuracy are both considered, and therefore the request scene of a large amount of request data can be efficiently and accurately judged.

Embodiments of the present application provide a fraudulent website detection method and related equipment. The fraudulent website detection method comprises: capturing a website to be detected, and extracting multi-dimensional website information; performing feature engineering on the multi-dimensional website information to obtain corresponding multi-dimensional website features; and inputting the multi-dimensional website features into a website detection model to obtain a detection result. Through the extraction and optimization of the multi-dimensional features of the website after capturing various website information, the present application provides multiple identification clues for the complex camouflage of fraudulent websites, further compresses redundant information, and strengthens key features related to fraud. These key features are input into the website detection model for multi-modal detection and classification, so as to efficiently and accurately determine whether the website is a fraudulent website. The present application not only can process complex multi-modal data, but also can quickly and efficiently identify fraudulent websites, improve detection accuracy, reduce misjudgment and omissions, and thus achieve the technical effects of improving user safety and reducing risks.

The invention relates to the technical field of AI vision, in particular to a method and device for intelligently detecting vehicle illegal driving through the AI vision technology and a storage medium. Compared with the prior art, the technical scheme provided by the invention has the following beneficial effects: the AI vision technology is adopted, efficient and accurate automatic identification and judgment of illegal driving behaviors are realized, the burden of manual auditing is reduced, the driving direction of the vehicle can be detected and the driving track of the vehicle can be tracked under the condition of limited resources, and the driving efficiency of the vehicle is improved. Therefore, the accuracy and the high efficiency of vehicle violation driving detection are improved.